Still having problems w/ definition of symbol 0

[Julia Harrell]

Hi Daniel

I think we are also experiencing this particular issue, but since we're MS4W
users, 
we don't have the ability to compile and use the latest source code. We did
upgrade 
to the MS4W 2.2.6 beta this morning, which uses MapServer 5.0-beta5. We were
hoping 
that your fix for this issue and the long-awaited deprecation of all the
various
hard-coded limits in ms-rfc-17 would be included this version and fix our
problems, 
but this "Undefined overlay symbol" error is still creating some havoc for
us.

The error message is:
---------------------

  ERROR	msLoadMap(): General error message. Undefined overlay symbol
'dashed5' 
  in class 3, style 0 of layer Geologic_Structures. msAddImageSymbol(): 
  Unable to access file. Error opening image file
C:\ms4w\Apache\cgi-bin/.\dashed5.

Problem symbol definitions in our symbol file start at:
----------------------------------------------------------
# Symbol Index 17

	SYMBOL
		NAME 'dashes'
		TYPE TRUETYPE
		FONT "Bookshelf Symbol 7"
		CHARACTER "u"
		ANTIALIAS TRUE
		GAP 24
		END
	END 

# Symbol Index 18
	
	SYMBOL
		NAME 'dashed5'
		TYPE ELLIPSE
		FILLED TRUE
		POINTS
			1 1
		END
		STYLE 0 12 4 16
		END
	END 	

The layer definition in our map file where we experience the problem:
----------------------------------------------------------------------

	LAYER
		NAME 'Geologic_Structures'
		GROUP 'Geology and Hydrogeology'
		DATA 'C:\gisdata\geol250s_arc.shp'
		PROJECTION
			"init=epsg:32119"
		END #end projection
		TYPE line
		STATUS DEFAULT
		TOLERANCE 8 #default is 3 for raster, 0 for vector
		
		CLASS
			NAME '1 - fault'
			EXPRESSION ('[TYPE]' eq '1')
			STYLE
				SYMBOL 'dashed2'
				COLOR 204 204 204
				SIZE 3
			END #end style
			STYLE
				SYMBOL 'dashed2'
				COLOR 051 051 051
				SIZE 2.5
			END #end style
		END # end class
		
		CLASS
			NAME '2 - fold'
			EXPRESSION ('[TYPE]' eq '2')
			STYLE
				SYMBOL 'dashed3'
				COLOR 051 051 051
				SIZE 2.5
			END #end style	
			STYLE
				SYMBOL 'dashed4'
				COLOR 051 051 051
				SIZE 8
			END #end style			
		END # end class
		
		CLASS
			NAME '3 - scarp'
			EXPRESSION ('[TYPE]' eq '3')
			STYLE
			        SYMBOL "circle"
				COLOR 051 051 051
                                SIZE 2.5                                	
                        END #end style
		END # end class
		
		CLASS # this all works, but not when using dashed5
			NAME '4 - geologic contact'
			EXPRESSION ('[TYPE]' eq '4')
			#STYLE
				SYMBOL 'dashed5'   ### tried SYMBOL 18 here too
				COLOR 051 051 051
				SIZE 2.5
			#END #end style
		END # end class
		
		CLASS # 28
			NAME '5 - structural symbol'
			EXPRESSION ('[TYPE]' eq '5')
			STYLE
				SYMBOL 'dashed3'
				COLOR 051 051 051
				SIZE 2.5
			END #end style
		END # end class
	END #end layer geologic structures 


In our mapfile, we have:

  47 layers in total

  111 classes in total, with a max of 21 in any given layer
        The layer we have issues with has only 5 classes

  112 styles in total, with a max of 21 in any given layer
        The layer we have issues with has distinct 7 styles in the 5
classes, 
        with the problem symbol being the sole symbol in its style & class

In our symbol file, we have 25 symbols, but we are currently only using 12
of them in
the map file. Symbol 18 ("dashed5") is giving us this particular error.
Symbols 1 - 16 all 
work, but symbols at index 17 or higher give us this same error if
referenced by name in 
the map file. If the symbold is referenced by index number, we don't get the
error and 
the map renders, but that particular symbol does not.

For the life of us, we can't see what's different with this symbol than most
of the 
others. Is there anything we can do to get around this problem, using
pre-compiled 
MS4W binaries? It's a real showstopper for us, because we have a lot of data
layers 
we really need to include in our groundwater quality decision support
viewer.

Best Regards,

Julia Harrell
GIS Coordinator, NC DENR
julia.harrell at ncmail.net



>>>The issue was happening only when accessing the first symbol in your 
>>>symbolset by name.
>>>
>>>I had accidentally screwed up the definition of symbol 0 (the default 
>>>symbol) in msInitSymbolSet(). This is fixed in r6339.
>>>
>>>Daniel
-- 
View this message in context: http://www.nabble.com/SVN-Broken-tf4103508.html#a12391657
Sent from the Mapserver - Dev mailing list archive at Nabble.com.